Lunar resource extraction

Also known as In-situ resource utilisation (ISRU), Lunar ISRU, Lunar ice mining

Lunar resource extraction means producing oxygen, water and metals from lunar regolith and ice, so that missions need to launch less material from Earth. Oxygen has been made from simulated lunar soil in Earth tests, and water has been detected remotely near the Moon's south pole, but no lunar extraction has yet been shown on the Moon.

Lab demonstrationDemonstratedReviewed 21 Oct 2026
AI illustration · Illustrative scene, not a photograph of the specific device or experiment.

In plain language

Everything sent from Earth to the Moon is expensive to launch and land. Lunar resource extraction, often called in-situ resource utilisation or ISRU, aims to make some of what missions need from material already on the Moon. The main targets are oxygen, which can be separated from the soil, water ice that may be present in craters near the poles, and metals that can be recovered from the soil.

So far the results sit in separate places. In 2025, NASA ran a test in a vacuum chamber on Earth that melted about 25 kg of simulated lunar soil and produced oxygen and metals. In 2009, a rocket stage was crashed into a crater at the Moon's south pole, and the debris plume was analysed for water. In 2025, a lunar drilling experiment did not reach the surface because its lander came to rest on its side. Moving from a test in a chamber to a working lunar supply needs many more steps.

Technical explanation

Two electrolysis routes have been tested with lunar soil simulants. Molten regolith electrolysis heats the soil until it melts, at about 1,700 °C in NASA's 2025 test, and produces oxygen and metals from the melt. The molten-salt route dissolves regolith in calcium chloride at about 950 °C. A University of Glasgow team, including PhD researcher Beth Lomax, reported that about 96% of the total oxygen was extracted from simulated regolith over about 50 hours, with about 75% released in the first 15 hours, leaving a mixed metal alloy. The method builds on the FCC molten-salt process, which Metalysis has scaled up for commercial metal and alloy production. These are laboratory results, and ESA's own description of the work says the process is still being adapted to the lunar context.

Water on the Moon has so far been detected remotely. On 9 October 2009 the LCROSS mission sent a spent Centaur rocket stage into the permanently shadowed Cabeus crater. Instruments on a companion spacecraft observed the ejecta plume and reported at most 155 ± 12 kg of water vapour and ice within the instrument's field of view. The authors estimated that water ice made up 5.6 ± 2.9% of the regolith by mass at the impact site. The water was inferred from spectral signatures rather than collected, and the concentration depends on an estimate of how much excavated regolith reached sunlight.

Turning these measurements into supply needs drilling, sample handling, processing, power and storage on the surface. PRIME-1 was designed to drill about three feet (roughly one metre) into the soil with its TRIDENT drill and to analyse samples for water and other components with the MSOLO mass spectrometer. The IM-2 lander came to rest roughly horizontal, which prevented the drill from reaching the surface, so the chain from excavation to on-site analysis was not completed. Processing, power and autonomous operation have not yet been tested together on the Moon.

What the evidence says

Each statement is typed by the strength of its evidence and linked to the source that supports it.

Demonstrated Shown in a published experiment, test or shipping product.

In a 2025 vacuum-chamber test at NASA's Kennedy Space Center on Earth, about 25 kg of simulated lunar soil was heated to about 1,700 °C and melted, and molecular oxygen and metals were produced from it.

Sources: [1] · Article body

In a laboratory molten-salt test on simulated lunar regolith at about 950 °C, about 96% of the total oxygen was extracted over about 50 hours, with about 75% released in the first 15 hours, leaving a mixed metal alloy.

Sources: [6] · Article body · medium confidence

The PDS context record for PRIME-1 states that the lander's final roughly horizontal orientation prevented the TRIDENT drill from reaching the lunar surface. NASA's mission page reports that the lander ended on its side and does not report any soil extraction or MSOLO analysis.

Sources: [3], [2] · PDS context record; mission page

On 9 October 2009, the LCROSS Centaur impact into the permanently shadowed Cabeus crater produced a plume in which instruments detected at most 155 ± 12 kg of water vapour and ice within the instrument's field of view.

Sources: [5] · Abstract

The LCROSS authors estimated that water ice made up 5.6 ± 2.9% of the regolith by mass at the impact site, using an estimate of the excavated regolith that reached sunlight. The water was inferred from spectral signatures, not sampled.

Sources: [5] · Abstract · medium confidence

In July 2024, NASA cancelled the VIPER rover after its cost grew past 30% of the original budget. NASA estimated savings of at least $84 million, and the Griffin lander mission continued with a mass simulator in place of the rover.

Sources: [4] · Article body

The oxygen extraction results reviewed here come from simulated lunar regolith in Earth tests. The sources checked do not show lunar regolith being used to produce oxygen or metals on the Moon.

Sources: [1], [6] · Article body

Proposed A design or plan that has not yet been built or tested.

NASA states that post-test data analysis is still needed, and that its reactor can be delivered to the Moon on future missions once the process has been perfected on Earth.

Sources: [1] · Article body

ESA's 2021 study set targets for a compact lunar oxygen payload: retrieve samples and extract 50 to 100 g of oxygen, extract more than 70% of the available oxygen from a lunar rock sample, and finish before sunset on the lunar surface (within fewer than 10 Earth days).

Sources: [7] · Article body · medium confidence

Limitations and open problems

  • The oxygen and metal results so far come from simulated lunar soil tested on Earth, in laboratories or vacuum chambers. Real lunar regolith may behave differently.
  • NASA says post-test data analysis is still needed, and that the reactor can be delivered to the Moon on future missions once the process has been perfected on Earth.
  • The 2025 PRIME-1 drill did not reach the lunar surface, so no regolith from depth was analysed on the Moon. The lander came to rest on its side.
  • Detecting water is not the same as mining it. The 5.6% ice estimate from LCROSS depends on an estimate of excavated mass and on spectral interpretation, and it does not establish a reserve that can be mined at a useful rate.
  • ESA's 2021 targets for a compact lunar oxygen payload are proposals. The sources checked here do not confirm a flight date or a result.
  • NASA cancelled the VIPER rover in July 2024 after cost growth passed 30% of its original budget, so the planned survey of polar ice did not go ahead.

Possible future uses

Speculative

Ideas, not predictions. None of these has been demonstrated.

  • Locally made oxygen for breathing and rocket propellant, which would reduce the mass launched from Earth if production proves reliable.
  • Metal alloys and sintered regolith used for landing pads, roads and berms, rather than for precision parts.
  • Water from polar ice used for drinking, life support and, after splitting, as a source of rocket propellant, if extraction and storage prove reliable.

Timeline

  1. 9 Oct 2009The LCROSS mission sends a spent Centaur rocket stage into the permanently shadowed Cabeus crater. The ejecta plume is later analysed for water. Sources: [5]
  2. 9 Oct 2019ESA reports a laboratory molten-salt test on simulated lunar regolith that extracts about 96% of the total oxygen over about 50 hours. Sources: [6]
  3. 29 Jul 2021ESA reports that industry teams are competing to design a compact lunar oxygen payload, with performance targets for a future lander. Sources: [7]
  4. Jul 2024NASA cancels the VIPER rover after cost growth passes 30% of its original budget. The Griffin lander mission continues, carrying a mass simulator in place of the rover. Sources: [4]
  5. 26 Feb 2025PRIME-1 launches on Intuitive Machines' IM-2 mission from NASA Kennedy. The lander later comes to rest roughly horizontal, and the drill does not reach the surface. Sources: [2], [3]
  6. 5 May 2025NASA reports a vacuum-chamber test at Kennedy Space Center in which molten regolith electrolysis produces molecular oxygen and metals from about 25 kg of simulated lunar soil. Sources: [1]

Futures this could help bring about

AI illustration
How far away?

Mining water and oxygen on the Moon

Engineering prototypes · 3 of 7 steps done

Glossary

ISRU
In-situ resource utilisation: collecting, processing, storing and using materials found at a destination instead of bringing them from Earth.
Regolith
The loose layer of broken rock and dust that covers the Moon's surface.
Simulant
A material made on Earth to copy the composition of lunar soil, used for testing.
Permanently shadowed region
A part of a crater, near a pole, that does not receive direct sunlight.

Sources

  1. government report
    NASA Kennedy Breathes Life into Moon Soil Testing
    NASA, 5 May 2025. Accessed 21 Oct 2026.
  2. government report
    Polar Resources Ice Mining Experiment 1 (PRIME-1)
    NASA, 11 Mar 2025. Accessed 21 Oct 2026.
  3. dataset
    CLPS Task Order PRIME-1 Athena Lander (NASA Planetary Data System context record)
    NASA Planetary Data System, Jul 2026. Accessed 21 Oct 2026.
  4. news
    NASA cancels half-billion dollar water-ice-seeking Moon rover
    Will Robinson-Smith. Spaceflight Now, 18 Jul 2024. Accessed 21 Oct 2026.
  5. peer reviewed paper
    Detection of Water in the LCROSS Ejecta Plume
    Colaprete, Schultz, Heldmann et al.. Science, 22 Oct 2010. DOI: 10.1126/science.1186986. Accessed 21 Oct 2026.
  6. government report
    Oxygen and metal from lunar regolith
    European Space Agency, 9 Oct 2019. Accessed 21 Oct 2026.
  7. other
    Competing companies developing payload to make air from moondust
    European Space Agency, 29 Jul 2021. Accessed 21 Oct 2026.

Revision history

3 changes since this profile was created.

Changes on 21 Oct 2026
Published on owner instruction (content loop iteration 10)d8d21cd
Sources and claims re-checked against originalsfbb434c
Initial profile drafted from primary sources7e66249